Hi Gang,
Am building up a new card, an arduino clone with 12 TPIC6B595 shift registers.
Trying to download a bootloader using IDE 1.0.1 with AVR ISP MKii.
Board is powered via FTDI Basic.
Seeing this error:
C:\Arduino-1.0.1\hardware/tools/avr/bin/avrdude -CC:\Arduino-1.0.1\hardware/tools/avr/etc/avrdude.conf -v -v -v -v -patmega328p -cstk500v2 -Pusb -e -Ulock:w:0x3F:m -Uefuse:w:0x05:m -Uhfuse:w:0xde:m -Ulfuse:w:0xff:m
avrdude: Version 5.11, compiled on Sep 2 2011 at 19:38:36
Copyright (c) 2000-2005 Brian Dean, http://www.bdmicro.com/
Copyright (c) 2007-2009 Joerg Wunsch
System wide configuration file is "C:\Arduino-1.0.1\hardware/tools/avr/etc/avrdude.conf"
Using Port : usb
Using Programmer : stk500v2
avrdude: usbdev_open(): did not find any USB device "usb"
Checked Vista Device Manager, AVR ISP is shown as working correctly under Jungo (not what I expected for sure!).
Is there a setting change I can make?
Thanks
This is what I remember. Hopefully, if it's moderately close to the truth, it will get you pointed in the right direction.
AVR ISP MKii has two drivers available. Only one driver can be installed at any given time. The Jungo driver is not compatible with avrdude. You have to remove the Jungo driver and install the (don't remember) driver.
I know that the AVR ISP MKii programmer and the Jungo driver have been discussed in the forum.
I had tried bootloading this same chip before using AVR Studio 5.1.
Just for a sanity check, I plugged the chip into a Duemilanove - and the L LED flashes 3 times slowly, so something must have gone on.
Burning bootloader with AVR ISP MKii selected as programmer gives the same error with the Duemilanove.
Just saw the Jungo driver note - will check that too.
Ok, a search for Jungo brought up a post by Diesel:
http://arduino.cc/forum/index.php/topic,110377.0.html
Windows complained a bit, but I got it installed - now to reboot the PC & see if it takes! ...
Okay, I'm back to report that uninstalling the Jungo driver and installing the driver found in the 1.0.1 folder that diesel pointed out above did the trick!
Can now download sketches! Got a simple one installed to blink an LED (on D8 tho, not 13), time to write something to show the shift registers work okay...
Yep. Designed to control strings of 12V LEDs used as segments of 7-segments displays.
Drive up to 12 digits at once.
Alltho it has occurred to me that with 96 bits available, a 9x9x9 cube would also be easy, using 81 bits for the cathode columns, and 9 more to control PNP or P-channel FETs for the anode layers. Just need more free time!
Here's another odd thing - why does it take about 20 seconds for this simple program to appear to start up?
I have 6 registers populated, all daisy chained, Serial.print of the data shows 1,2,4,8,10,20,40,80 being generated correctly right after bootup.
Yet 20-25 seconds before I see an LED/resistor from +5 to any output pin start flashing.
#include <SPI.h>
byte ledpin1 = 8;
byte SH_CLK = 10;
void setup(){
pinMode (ledpin1, OUTPUT);
pinMode (SH_CLK, OUTPUT);
digitalWrite (SH_CLK, HIGH);
Serial.begin(9600);
SPI.begin();
}
void loop(){
for (int x = 1 ; x<0x81; x=x<<1){
digitalWrite (SH_CLK, LOW);
SPI.transfer (x);
SPI.transfer (x);
SPI.transfer (x);
SPI.transfer (x);
SPI.transfer (x);
SPI.transfer (x);
digitalWrite (SH_CLK, HIGH);
delay (50);
Serial.println(x,HEX);
}
}
will check back tomorrow ... good night ...
These...
pinMode (SH_CLK, OUTPUT);
digitalWrite (SH_CLK, HIGH);
...should probably be swapped...
digitalWrite (SH_CLK, HIGH);
pinMode (SH_CLK, OUTPUT);
Try setting the DIM pin LOW.
Thanks. Home from fencing, I was just describing the board operation to my wife, and it dawned on me I hadn't done that.
If I swap the other two, won't that turn on the pullup resistor, then set the pin to an output but not put it at any particular state?
I wanted it set to an output and then commanded high.
Yes, it does first turn on the pullup resistor. After switching the pin to an output the final state is HIGH OUTPUT.
It makes a difference if you have an external pullup resistor on the line (or if the shift register has an internal pullup). If you do it the other way (pinMode first), there will be short low pulse when the pin is first changed to an output.
I can't see how it will help with the 20 second problem (though I've seen stranger things). It may eliminate a startup glitch and it certainly won't hurt anything.
